Skip to navigation
Image Series

Gemini Image Models

View as Markdown

Gemini image models use the image generation endpoint with a Generate Content-compatible request body. Use gemini-3.1-flash-image for general image generation, gemini-3-pro-image for professional visual assets, and gemini-2.5-flash-image for existing Nano Banana workflows.

curl https://api.getopenmodels.com/v1/images \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OM_API_KEY" \
  -d '{
    "model": "gemini-3.1-flash-image",
    "route": { "provider": "<provider-for-this-model>" },
    "route_mode": "balanced",
    "contents": [
      {
        "role": "user",
        "parts": [
          {
            "text": "Generate an image of a Chinese Li Hua cat in a modern studio product-photo style."
          }
        ]
      }
    ],
    "generationConfig": {
      "responseModalities": ["TEXT", "IMAGE"],
      "responseFormat": {
        "image": {
          "aspectRatio": "16:9",
          "imageSize": "2K"
        }
      }
    }
  }'

For Gemini image models, set image generation options in generationConfig.

FieldTypeNotes
modelstringSupported values: gemini-3-pro-image, gemini-3.1-flash-image, gemini-2.5-flash-image.
contentsarrayRequired Generate Content input. Use user messages with parts for text prompts and optional image inputs.
contents[].parts[].textstringText prompt that describes the image to generate or edit.
contents[].parts[].inline_dataobjectOptional image input for reference-image or image-editing workflows. Set mime_type and base64 data.
generationConfig.responseModalitiesarrayOutput modalities. Use ["TEXT", "IMAGE"] when you want both text and image parts, or ["IMAGE"] for image-only output.
generationConfig.responseFormat.image.aspectRatiostringOutput aspect ratio. Common values include 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9. gemini-3.1-flash-image also supports wider ratios such as 1:4, 4:1, 1:8, and 8:1.
generationConfig.responseFormat.image.imageSizestringOutput size for gemini-3-pro-image and gemini-3.1-flash-image. Valid values include 1K, 2K, and 4K; gemini-3.1-flash-image also supports 512. Omit this field for gemini-2.5-flash-image.
tools[].google_searchobjectOptional grounding tool for Gemini 3 image models when the prompt needs current or factual web information.

Gemini image models also support OpenAI-compatible multimodal chat messages. Use this shape when your client sends image inputs through messages[].content[]:

curl https://api.getopenmodels.com/v1/chat/completions \
  -H "Authorization: Bearer $OM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3.1-flash-image",
    "route": { "provider": "<provider-for-this-model>" },
    "route_mode": "balanced",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "text",
            "text": "What is in this image?"
          },
          {
            "type": "image_url",
            "image_url": {
              "url": "https://live.staticflickr.com/3851/14825276609_098cac593d_b.jpg"
            }
          }
        ]
      }
    ]
  }'